On this appeal from a judgment entered after a bench trial in the Southern District of New York, Richard Owen, District Judge, 407 F.Supp. 1370, the essential question presented is whether the work performed respectively by the heavy and light cleaners employed by Columbia University is “equal” within the meaning of the Equal Pay Act of 1963, 29 U.S.C. § 206(d)(1) (1970)1 (the Act). We agree with the district court that the work is not equal. We affirm.2
I.
The Equal Pay Act prohibits an employer from discriminating “between employees on the basis of sex by paying wages to employees . . . at a rate less than the rate at which he pays wages to employees of the opposite sex . . . for equal work on jobs the performance of which requires equal skill, effort, and responsibility, and which are performed under similar working conditions . . . . ” See note 1 supra.
Invoking the enforcement provisions of the Fair Labor Standards Act,3 the Secretary of Labor commenced the instant action against Columbia University and its president (Columbia) on March 11, 1974. The action sought to enjoin Columbia from discriminating against its female light cleaners, allegedly on the basis of sex in violation *956of the Equal Pay Act, by paying them at a lower hourly rate than that paid to its male heavy cleaners. The action also sought an injunction against further violations and an award of back pay.
At a 15 day bench trial during September and November 1974, the parties, focusing upon the statute’s “equal . . . effort” criterion, produced extensive evidence on the physical requirements of the heavy and light cleaning positions. The crux of Judge Owen’s finding of fact on this critical issue is set forth in his opinion dated February 11, 1976 as follows:
“[T]he jobs of heavy cleaner and light cleaner are different. Going beyond the job descriptions, from the extensive testimony adduced before me at the trial, it is clear that the job of heavy cleaner involves greater effort than that of light cleaner.” 407 F.Supp. at 1374-75.
Accordingly, the judge held that the Secretary had failed to sustain the burden of proving that the heavy and light cleaners perform equal work within the meaning of the Act. From the judgment dismissing the action, the Secretary has taken the instant appeal.4
II.
The facts are not in dispute. They appear clearly on the face of the record. To the extent necessary to an understanding of our rulings on the issues of law presented, we summarize the controlling facts, including the essential findings of the district court which we accept pursuant to Fed.R. Civ.P. 52(a).
For more than 30 years Columbia has divided the duties of its custodial force into “heavy” and “light” categories. At the time of trial it employed 160 heavy cleaners (designated “janitors” prior to 1972), 4 of whom were female, and 111 light cleaners (designated “maids” prior to 1972), all of whom were female. Since 1972 all heavy cleaning positions have been open to applicants of both sexes.5 No male ever has applied for the position of light cleaner. The light cleaners always have been paid less than the heavy cleaners. The differential was 45 cents an hour at the time of trial.
Columbia assigns approximately 80% of the light cleaners to its academic buildings and the remainder to its residence halls. Each cleaner in the group assigned to the academic buildings has daily responsibility for between 1 and 5 floors, depending on the size of the building and the functions of the particular rooms included in the assignment. The functions are those one would expect at any university, including classrooms, offices, library stacks, and the like. The light cleaners dust mop or vacuum the floors of the rooms and some of the corridors; dust, polish, or dampcloth the furniture, fixtures, baseboards and windowsills; and remove small spots from the walls, floors and doors. They also empty wastebaskets and ashtrays into trashbags. These are deposited at the elevator on each floor where they are picked up by heavy cleaners. In performing their work the light cleaners use dust mops, carpet sweepers, household vacuum cleaners, toy brooms, rags, sponges and 14 quart buckets. Some of the light *957cleaners transport this equipment in wheeled carts which weigh 60 pounds unloaded. The equipment itself, including a vacuum cleaner, creates a 21 pound load, to which a full trash bag adds an indeterminate amount of weight up to 30 pounds.
The light cleaners assigned to the residence halls perform the same functions as those performed in the academic buildings. Each day they clean the common rooms and offices on the ground floors, and the corridors and lounges on the residential floors. Students’ rooms are cleaned on a special assignment basis when unoccupied during the summer.
The heavy cleaners may be grouped roughly into four categories: (1) those assigned to public corridors, lobbies, stairways and elevators; (2) those assigned to public lavatories; (3) those permanently assigned to special projects; and (4) those assigned to certain off-campus buildings.
The primary daily responsibility of group (1) is to wet mop the corridors, lobbies, stairways, elevators, classrooms and laboratories of the academic buildings. They also collect the bags of trash left on each floor by the light cleaners and transport them to dumping areas.
The heavy cleaners in group (2) clean daily the university’s public lavatories. They dust and wet mop the floors, clean the toilets and sinks, empty the trash, and wash marks from the walls. Many of the heavy cleaners in this group assist the heavy cleaners in the first group with the job of removing trash. Depending on the load this takes from less than one-half hour to 2 hours daily.
In their wet mopping both of these groups use a dolly which carries two 44 quart buckets with wringers and cleaning solution. The dolly weighs 60 pounds empty and up to 140 pounds full. For larger areas they use mop trucks which hold 30 or 60 gallons. These weigh from 160 to 200 pounds empty and from 300 to 500 pounds full. Trash removal is accomplished by the use of trucks. These are 6 feet long, 3 feet wide and 6 feet high, and weigh 200 pounds empty and up to 800 pounds full.
The heavy cleaners in groups (1) and (2) perform additional tasks on an occasional basis. These include loading, unloading and transporting drums of cleaning solvent which hold 55 gallons of liquid and weigh from 400 to 600 pounds; climbing 14 foot ladders to clean in high places and change light bulbs; vacuuming carpets with industrial machines which weigh from 74 to 96 pounds empty and from 122 to 192 pounds full; cleaning carpets with motorized pile lifters; shovelling snow from entrance ways; and turning on and off heavy ventilation equipment.
The special projects handled by group (3) of the heavy cleaners are jobs done regularly but infrequently. These include scrubbing and stripping floors, shampooing rugs and washing Venetian blinds.
The work assignments of those classified in groups (1), (2) and (3) are not necessarily hard and fast. For example, in Avery Hall, one of the smaller buildings, one heavy cleaner wet mops all the corridors and staircases, cleans the lavatories and does any project work. (Two light cleaners handle the offices and libraries in that building). In contrast, in a large building such as Butler Library, assignments of the 5 heavy cleaners in the building are divided among the 3 classifications. (Eleven light cleaners are assigned to Butler).6
The heavy cleaners in group (4) are assigned to small converted office buildings located on streets adjacent to Columbia’s main campus. They primarily perform light cleaning duties. Light cleaners handled these assignments until 1974 when considerations of security prompted Columbia to substitute male employees. The buildings involved are outside of the security perimeter of the main campus and are cleaned during the midnight to 8 A.M. shift. *958Before the change heavy cleaners escorted the light cleaners from the main campus to these buildings and locked them in. Each light cleaner then spent the night alone in her respective building and later was escorted back to the main campus.
These off-campus buildings are the only exception to the otherwise uniform line of demarcation which Columbia maintains between “heavy” and “light” cleaning tasks. Written job descriptions listing the heavy and light cleaning tasks are incorporated in the collective bargaining agreements between Columbia and the Transport Workers’ Union.7
The cleaners themselves are well aware of the distinction between heavy and light assignments and have not hesitated to enforce it. In 1971, for example, 8 light cleaners successfully protested an assignment to remove heavy trash. Columbia moreover has changed from “light” to “heavy” certain assignments which technically were in the “light” category but which in practice called for physical exertion comparable to that of the heavy cleaning tasks. The wastebaskets in the Journalism building, for example, frequently are filled with books, magazines and other heavy materials. After the light cleaners in the building complained of the resulting burden, the job of emptying these wastebaskets was reassigned to heavy cleaners. The heavy cleaners also collect the heavy trash which accumulates in the chemistry laboratories and the architecture drafting rooms.
The light cleaners’ perception of the distinction between their duties and those of the heavy cleaners has manifested itself in other ways as well. Seven light cleaners applied for and were accepted as heavy cleaners when Columbia first opened the heavy cleaner category to women in 1972. On-the-job training commenced. By the end of 7 weeks, 4 of the 7 had transferred back to light cleaning.8 No light cleaners have applied for heavy cleaner openings posted since that time. At trial several light cleaners testified that they had not applied for heavy cleaning positions because heavy cleaning tasks were beyond their capacity. As one of them put it, “I cannot do a heavier job than I have already.”
There of course was much other evidence adduced at the 15 day trial, as appears in the 4 volume trial transcript. The foregoing summary is believed to be sufficient, however, for an understanding of our rulings below, especially in light of the district court’s clear, concise and adequate findings of fact based on the undisputed evidence.
III.
Since job content is a matter determined by the particular employer, whether two job classifications entail “equal work” under the Act necessarily must be decided on a case-by-case basis. Brennan v. Prince William Hospital Corp., 503 F.2d 282, 286 (4 Cir. 1974), cert. denied, 420 U.S. 972 (1975); Hodgson v. Golden Isles Convalescent Homes, Inc., 468 F.2d 1256, 1258 (5 Cir. 1972) (per curiam). The burden of proving that “equal work” is involved is on the Secretary. Corning Glass Works v. Brennan, 417 U.S. 188, 195 (1974), aff’g Hodgson v. Corning Glass Works, 474 F.2d 226 (2 Cir. 1973). But in meeting that burden the Secretary does not have to prove that the duties performed are identical. A violation of the Act may be proven if the “skill, effort, and responsibility” required in the performance of the jobs is “substantially equal.” Hodgson v. Corning Glass Works, supra, 474 F.2d at 234; Schultz v. Wheaton Glass Co., 421 F.2d 259, 265 (3 Cir.), cert. denied, 398 U.S. 905 (1970). One court has suggested a balancing function in applying this “substantially equal” test:
“[Congress] did not authorize the Secretary or the courts to engage in wholesale reevaluation of any employer’s pay structure in order to enforce their own concep*959tions of economic worth .... But if courts defer to overly nice distinctions in job content, employers may evade the Act at will.” Brennan v. Prince William Hospital Corp., supra, 503 F.2d at 285.
The “equal effort” criterion — the principal issue in the instant case — has been elaborated upon extensively in prior cases and in the interpretive bulletin promulgated by the Secretary. 29 C.F.R. §§ 800.114-800.163 (1976).9 Under the Act, “effort” is the physical or mental exertion required in performing a job. So long as the ultimate degree of exertion remains comparable, the mere fact that two jobs call for effort different in kind will not render them unequal. Brennan v. South Davis Community Hospital, 538 F.2d 859, 863-64 (10 Cir. 1976); Brennan v. Board of Education, Jersey City, New Jersey, 374 F.Supp. 817, 827 (D.N.J. 1974); 29 C.F.R. § 800.127 (1976). Nor will effort expended on additional tasks assigned to male employees necessarily suffice to justify a pay differential. If the additional tasks do not consume a significant total amount of all of the employees’ time, Hodgson v. Behrens Drug Co., 475 F.2d 1041, 1049 (5 Cir.), cert. denied, 414 U.S. 822 (1973); Hodgson v. Brookhaven General Hospital, 436 F.2d 719, 725 (5 Cir. 1970); or if female employees also perform duties which require additional effort, Brennan v. Prince William Hospital Corp., supra, 503 F.2d at 286; or if third persons who perform the additional tasks as their primary job are paid less than the male employees in question, see Shultz v. Wheaton Glass Co., supra, 421 F.2d at 266 — in these situations the additional effort is insufficient to differentiate the male positions under the Act.
These rules, formulated to embrace situations where differences in effort are merely illusory, Ross and McDermott, The Equal Pay Act of 1963: A Decade of Enforcement, 16 B.C.Ind. & Com.L.Rev. 1, 38 (1974), significantly limit the weight which should be given to the fact that male and female employees who perform similar work in fact perform different duties. The Secretary contends on the facts of this case that the application of these rules requires reversal. His argument has two prongs. First, he argues that, notwithstanding the differences between the particular tasks performed by the heavy and light cleaners, they ultimately expend equivalent amounts of effort. What the light cleaners save in not having to deal with heavy equipment, so the argument goes, they expend in having to cover larger areas and in handling additional fixtures and furniture.10 Second, he attacks the significance the district court attached to the heavy cleaners’ heavier equipment. According to the Secretary, there is “nothing in the record ... to support the court’s assumption that the [heavy cleaners’] equipment . . . which was either used for infrequent periods or in combination with other [heavy cleaners] or was on wheels and could be moved easily, required any greater effort.”
Turning to the first prong, it is true that the light cleaners cover a great deal of ground each night. But the evidence of distribution of work assignments among the buildings shows that the amount of ground covered by the heavy cleaners, while less extensive by some number of square feet not specified by the Secretary, cannot be dismissed so easily. The very heavy cleaners the Secretary singles out as responsible for small areas testified to difficulties in completing their daily assignments. Moreover the Secretary’s own expert witness testified that the heavy cleaners’ corridors, *960lobbies, stairways, elevators and lavatories collect a greater concentration of dirt than do the light cleaners’ offices and classrooms and hence require greater cleaning effort.
As for the second prong, we believe that the district court attached appropriate significance to the heavy cleaners’ equipment. The record clearly supports the inference that the heavy cleaners’ principal task of wet mopping heavily traversed areas requires greater physical effort than wielding dust mops and rags. This is true not only because the surfaces are dirtier but also because the equipment is substantially bulkier and heavier.
The concept of “effort” in the Act is straightforward. It calls for a direct comparison of the amount of physical exertion required by the jobs; there is no factor added to compensate for physiological differences between men and women. Based on our careful review of the record before us we cannot say that the district court was clearly erroneous in making this direct comparison and in finding as a fact that heavy cleaning involves “greater effort”.
We are mindful of other cases which, in applying the Act to janitors and maids who perform the same general cleaning duties as Columbia’s heavy and light cleaners, have held that the jobs involve substantially equal effort. Brennan v. South Davis Community Hospital, supra, 538 F.2d at 863-64; Brennan v. Goose Creek Consolidated Independent School District, 519 F.2d 53, 58 (5 Cir. 1975); Brennan v. Board of Education, Jersey City, New Jersey, supra, 374 F.Supp. at 828-29; but see Marshall v. Marist College, No. 74 Civ. 4713-LWP (S.D.N.Y. June 29, 1977.) Close scrutiny of the facts of these cases, however, shows that the division of duties in the instant case is materially different. The cited cases follow the fact pattern which has emerged as the dominant one in cases under the Act involving all types of work. The central fact is the sharing of common responsibilities by male and female workers. The most frequently litigated question is whether additional lifting, fetching, hauling or other tasks performed by men require an amount of additional effort sufficient to foreclose a holding of substantial equality under the Act. For the most part the cases have concluded that the additional duties are either too insubstantial in amount or too inconsistently assigned to measure up to the Act’s standard of substantial equality.11 See also Hodgson v. Corning Glass Works, supra, 474 F.2d at 234; Peltier v. City of Fargo, 533 F.2d 374, 376-79 (8 Cir. 1976); Brennan v. Prince William Hospital Corp., supra, 503 F.2d at 286-90; Hodgson v. Behrens Drug Co., supra, 475 F.2d at 1048-50, 1051; Hodgson v. Golden Isles Convalescent Homes, Inc., supra, 468 F.2d at 1257-58; Schultz v. American Can Co.-Dixie Products, 424 F.2d 356, 360-61 (8 Cir. 1970); Schultz v. Wheaton Glass Co., supra, 421 F.2d at 262-63.
In the instant case, by contrast, the male employees’ duties consistently require *961greater effort. Columbia’s job classifications cannot be equated with the makeweight classifications so frequently found in the prior cases. See Angelo v. Bacharach Instrument Co., 555 F.2d 1164, 1172 & n. 11 (3 Cir. 1977). Moreover, since the “additional” duties of Columbia’s heavy cleaners in fact comprise their entire jobs, the increment of effort involved is more “substantial” than that involved in any of the cases cited above.
Both management and employees at Columbia for more than 30 years have recognized the existence of a material difference in the amount of effort required by heavy and light cleaning. The job classifications here involved never were entirely sex-based. Ever since 1949, long before most heavy cleaning positions were opened to women on an equal basis, Columbia has paid heavy cleaners’ wages to women employed in tasks equivalent to those performed by the male cleaners. See note 5 supra. The record shows that the light cleaners are cognizant of the difference. It supports a finding that even today it is the additional effort demanded of heavy cleaners which has deterred the vast majority of Columbia’s light cleaners from availing themselves of the opportunity to eliminate the 45 cent hourly disparity in pay.
Based on this evidence of the understanding and experience of those most closely involved, together with the undisputed fact that heavy cleaning calls for greater effort, we hold, as the district court did, that heavy and light cleaning are not substantially equal work within the meaning of the Act.
In reaching this conclusion we have excluded from consideration the comparatively small number of heavy cleaners assigned to off-campus office buildings who perform work identical to that performed by the light cleaners on campus. On the face of it, this recently created category of heavy cleaners would fall within the frequently cited rule mentioned above that additional tasks must “consume a significant amount of the time of all those whose pay differentials are to be justified in terms of them. . . .” Hodgson v. Brookhaven General Hospital, supra, 436 F.2d at 725. See also 29 C.F.R. § 800.128 (1976).
But assuming that this strict reading of the statute applies to the heavy cleaners assigned to off-campus buildings, even though the majority of the employees in the “male” classification perform work consistently demanding greater effort than that required in the “female” classification, an additional factor precludes its application here. Judge Owen found that the light cleaning in question is performed “alone at night in off-campus buildings located outside the University’s security perimeter.” Taken in the context of his opinion, this necessarily implies a further finding that the off-campus work is performed in more dangerous conditions. The record does show that solicitude for the safety of the light cleaners motivated Columbia to change the assignments.
These circumstances call for the application of the provision of the Act which permits pay differentials between work performed under dissimilar “working conditions”. As defined by the Supreme Court in Corning Glass Works v. Brennan, supra, 417 U.S. at 202, this concept encompasses the “hazards” of employment. Specifically, “the physical hazards regularly encountered, their frequency, and the severity of injury they can cause” are taken into account. Id.; see Hodgson v. Corning Glass Works, supra, 474 F.2d at 231. This case does not involve the dangers of a machine on an assembly line, but the risk of exposure to urban crime faced by women working alone and unguarded in the middle of the night. So long as the hazard is a genuine one, however, its source is not determinative under the Act.
True, the record here does not show that any light cleaner, on or off campus, has been a victim of crime while on the job. One court has demanded such tangible proof of a hazard’s existence before applying the “working conditions” provision of the Act. See Hodgson v. Daisy Manufacturing Co., 317 F.Supp. 538, 543-44 (W.D. Ark.1970), aff’d per curiam, 445 F.2d 823 *962(8 Cir. 1971). In the instant case however we hold that the imposition of such a threshold requirement would be unreasonable. Columbia, a great university forced to cope with the sometimes uncongenial contemporary urban environment, is entitled to recognition of the preventive measures it takes. Moreover the university’s good faith in this matter is not open to question. Only genuine concern could have motivated it to increase its custodial costs and, barely two months before the instant action was commenced, risk an adverse legal impact. Indeed, the Secretary has described Columbia’s action in reassigning these buildings to the heavy cleaners as “laudable”.
Nor do we attach significance to the fact that the heavy cleaners have not had these off-campus assignments during the indefinite past. Any other interpretation of the Act would deter employers who maintain job classifications in compliance with the Act from taking corrective action once they conclude that some existing security or safety arrangement is unsatisfactory. We decline to sanction any such interpretation of the Act.
Affirmed.
. 29 U.S.C. § 206(d)(1) (1970) provides:
“No employer having employees subject to any provisions of this section shall discriminate, within any establishment in which such employees are employed, between employees on the basis of sex by paying wages to employees in such establishment at a rate less than the rate at which he pays wages to employees of the opposite sex in such establishment for equal work on jobs the performance of which requires equal skill, effort, and responsibility, and which are performed under similar working conditions, except where such payment is made pursuant to (i) a seniority system; (ii) a merit system; (iii) a system which measures earnings by quantity or quality of production; or (iv) a differential based on any other factor other than sex: Provided, That an employer who is paying a wage rate differential in violation of this subsection shall not, in order to comply with the provisions of this subsection, reduce the wage rate of any employee.”
. This is our second encounter with the Equal Pay Act. Our first is reflected in Chief Judge Friendly’s helpful opinion in Hodgson v. Corning Glass Works, 474 F.2d 226 (2 Cir. 1973), aff’d sub nom. Corning Glass Works v. Brennan, 417 U.S. 188 (1974), which of course is the law of this Circuit.
. 29 U.S.C. § 217 (1970).
. Judge Owen also dismissed a companion action, Walker v. Columbia, 73 Civ. 2687, which had been consolidated for trial with the instant action. The Walker action had been brought pursuant to Title VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e-2000e-17 (1970), by individual women employees of Columbia who alleged discrimination in hiring, job placement, pay, promotions and training. The notice of appeal from the judgment dismissing the Walker action was withdrawn.
. For heavy cleaning work Columbia has been paying more in cleaning wages to a limited number of women since 1949 when the classification “shower maid” first appeared in its collective bargaining agreement. This classification encompassed the female cleaners assigned to the showers and lavatories in the women’s residence hall at Columbia. The classification never included more than 8 employees. At the time of trial 2 of the 4 female heavy cleaners performed the “shower maid” function.
All other heavy cleaner positions were limited to men until 1972 when Columbia opened them to women. The remaining 2 women employed as heavy cleaners at the time of the trial worked within this board, formerly male, classification.
. Heavy cleaners predominate in some of the buildings. For example, Chandler, the chemistry building, has 3 heavy cleaners and 2 light cleaners. One heavy cleaner mops the laboratories and corridors; the second cleans the lavatories and removes the trash; the third does special project work. The 2 light cleaners handle the offices and library in the building.
. The Transport Workers’ Union is the exclusive bargaining agent for all of Columbia’s custodial and maintenance employees.
. Two of the group remain as heavy cleaners. The seventh has transferred to a position outside of the buildings and grounds department.
. Other courts have used the Secretary’s interpretation as an aid in applying the Act. E. g., Brennan v. Owensboro-Daviess County Hospital, 523 F.2d 1013, 1028 & n. 13 (6 Cir. 1975).
. The mainstay of the Secretary’s argument in this respect is a comparison between the daily responsibilities of two light cleaners and those of two heavy cleaners assigned to the public lavatories in the same academic building. Specifically, a light cleaner responsible for 60 library “cubicles”, 8 carpeted offices, 5 other rooms, a classroom, a private toilet, and a flight of stairs, is compared with a heavy cleaner responsible for the 16 public lavatories, ranging in size from 1 toilet and sink to 4 toilets and sinks, and the trash detail in a building used by 10,000 people daily.
. In Brennan v. South Davis Community Hospital, supra, the janitors and maids performed common duties, except that the janitors (1) occasionally operated a floor stripping machine; (2) refilled a soft drink machine; (3) carried trash cans to an outside receptacle; (4) used ladders; and (5) shovelled snow. The maids also worked on ladders, occasionally carried trash, and had additional cleaning duties not shared with the janitors. The Tenth Circuit, while recognizing that the janitors’ additional duties did require “extra effort”, found the amount too insubstantial to preclude the Act’s application. 538 F.2d at 863-64.
Brennan v. Goose Creek Consolidated Independent School District, supra, was to the same effect. There the employer’s job description somewhat resembled the actual division of duties at Columbia. But in practice the janitors and maids did almost identical work. 519 F.2d at 58.
In Brennan v. Board of Education, Jersey City, New Jersey, supra, the maids and janitors both performed the basic wet mopping, sweeping, and dusting in classrooms, corridors and lavatories. The court found the janitors’ yard work, snow removal, ladder work, light hauling and floor stripping either too incidental to overcome the basis in “common chores [performed] for the bulk of the workday” or balanced by additional tasks assigned to the maids. 374 F.Supp. at 828-29.
See also Hodgson v. San Diego Unified School District, 71 CCH Lab. Cas. ¶ 32,390 (S.D.Cal.1973).